TPFDF - Performance Measurement

TPFUG_Req_Number: D94004
Contact Originator Id: Jelmar Blauw
Abstract: Improvements in the TPFDF Data Collection software and reporting
Description: A TPFDF Data Collection tool is incorporated in the TPFDF package. It measures the usage of TPFDF Macros and associated SVCs. The tool allows two types of displays. The first display is by File-ID (ZUDFM OAM/STA/file_id). This will display all activities for TPFDF Macros and SVCs during the latest collection. The second display is by Macro/SVC-number (ZUDFM OAM/GRAPH/entry_no/file_id). This will display all historical data for that particular Macro/SVC for a particular file_id.

As development increase within the TPFDF environment, and efficiency becomes more of a concern, a need for a better efficiency-monitoring tool becomes apparent. Although the current TPFDF Data Collection tool is very useful, it lacks comprehensive reporting features.

A reporting facility is required, which highlights potential problem areas, based on the figures in the most recent measurement-runs. In order to identify trends, a better report of previous runs is required.

This can be achieved in a TPF-environment, but preferably it will be downloaded to another platform (PC, MVS), where provided reporting programs (C, SAS) can run. The DBA or Performance Engineer must be able to create relationships and set thresholds, e.g. when the ratio of CFIND:PFIND > x a warning should be issued in the report.

Currently, the ZUREC and Recoup statistics are available in the same file structure as the current TPFDF Data Collection Data. Ideally, the new tool would incorporate the ZUREC and Recoup data as well as the TPFDF Data Collection Data. Enabling the tool to include these statistics, allows the reports to be evaluated against size and volume.

Also, a historical display of Recoup information would be helpful in identifying database growth. The DBA or Performance Engineer must be able to set thresholds so that warnings can be issued when databases grow over the threshold.

A TPF system should not be impacted by the addition of reporting programs. Therefore, it is preferred that the interpretation of the collected data will not be performed on a TPF-system.
Critical Success Factors: The new tool will have to retain historical data and the ability to display this historical data. This display needs to be based on the user defined relationships and thresholds.
Alternative Solutions: Improve the display of the current tool to include user specified relationships and thresholds. This will lack the required reporting facilities to identify trends.
Solution Considerations: The solution should be implemented in such a manner that users can write additional reporting programs to fulfil their specific needs.